Rabies transmitted by domestic dogs is a serious yet neglected public health threat in many underserved communities in Africa and Asia. Achieving 70% vaccination coverage in dog populations through annual mass vaccination campaigns is an effective means of controlling the disease in these communities. Evaluating the extent to which this target coverage is achieved requires either accurate pre-campaign estimates of the dog population size or accurate estimates of the coverage attained by conducting post-vaccination surveys. Short-term marking of dogs by applying visible marks may be useful to achieve these estimates, but will be affected by the performance of the marking methods. We evaluated the longevity and visibility of two readily-available livestock marking methods applied to owned, free roaming dogs.We applied two types of marks (spray and crayon) with three different colours (red, blue and green) to each of 21 dogs and compared the time of persistence of the marks over several weeks. Two independent observers assessed the visibility and colour of the marks. Each dog was observed over 2 eight to 37 days (median: 28 days). Kaplan-Meier survival analyses and semi-parametric log-rank tests were performed separately for both observers. Spray marks remained visible significantly longer (median of 24 days for both observers) compared with crayon marks (medians of 10 and 13 days). After 10 days, 90% of spray marks were still visible, compared with only 46% of crayon marks.Visibility of marks was reduced in darker-coloured dogs. Colours of marks were frequently misclassified, and agreement between observers on the colours of the marks was low (Cohen's kappa coefficient = 0.27).The livestock marker spray can effectively be used to mark dogs that are physically restrained, for example during vaccination campaigns. Resight surveys should be conducted within a short a time as possible after marking; however, our results suggest that loss of marks will not have a significant impact if surveys are conducted within 5-7 days after marking with the spray. Results that depend on observers' abilities to distinguish between the three colours which we evaluated may not be reliable.